Post on 24-Nov-2021
transcript
tinyML Talks Sponsors
Additional Sponsorships available – contact Bette@tinyML.org for info
tinyML Strategic Partner
3 © 2020 Arm Limited (or its affiliates)3 © 2020 Arm Limited (or its affiliates)
Optimized models for embedded
Application
Runtime(e.g. TensorFlow Lite Micro)
Optimized low-level NN libraries(i.e. CMSIS-NN)
Arm Cortex-M CPUs and microNPUs
Profiling and debugging
tooling such as Arm Keil MDK
Connect to high-level
frameworks
1
Supported byend-to-end tooling
2
2
RTOS such as Mbed OS
Connect toRuntime
3
3
Arm: The Software and Hardware Foundation for tinyML
1
AI Ecosystem Partners
Resources: developer.arm.com/solutions/machine-learning-on-arm
Stay Connected
@ArmSoftwareDevelopers
@ArmSoftwareDev
PAGE 4| Confidential Presentation ©2020 Deeplite, All Rights Reserved
BECOME BETA USER bit.ly/testdeeplite
WE USE AI TO MAKE OTHER AI FASTER, SMALLER AND MORE POWER EFFICIENT
Automatically compress SOTA models like MobileNet to <200KB with
little to no drop in accuracy for inference on resource-limited MCUs
Reduce model optimization trial & error from weeks to days using
Deeplite's design space exploration
Deploy more models to your device without sacrificing performance or
battery life with our easy-to-use software
Copyright © EdgeImpulse Inc.
TinyML for all developers
Get your free account at http://edgeimpulse.com
Test
Edge Device Impulse
Dataset
Embedded and edge
compute deployment
options
Acquire valuable
training data securely
Test impulse with
real-time device
data flows
Enrich data and train
ML algorithms
Real sensors in real time
Open source SDK
Maxim Integrated: Enabling Edge Intelligencewww.maximintegrated.com/ai
Sensors and Signal Conditioning
Health sensors measure PPG and ECG signals critical to understanding vital signs. Signal chain products enable measuring even the most sensitive signals.
Low Power Cortex M4 Micros
The biggest (3MB flash and 1MB SRAM) and the smallest (256KB flash and 96KB SRAM) Cortex M4 microcontrollers enable algorithms and neural networks to run at wearable power levels
Advanced AI Acceleration
The new MAX78000 implements AI inferences at over 100x lower energy than other embedded options. Now the edge can see and hear like never before.
Wide range of ML methods: GBM, XGBoost, Random
Forest, Logistic Regression, Decision Tree, SVM, CNN, RNN,
CRNN, ANN, Local Outlier Factor, and Isolation Forest
Easy-to-use interface for labeling, recording, validating, and
visualizing time-series sensor data
On-device inference optimized for low latency, low power
consumption, and a small memory footprint
Supports Arm® Cortex™- M0 to M4 class MCUs
Automates complex and labor-intensive processes of a
typical ML workflow – no coding or ML expertise required!
Industrial Predictive Maintenance
Smart Home
Wearables
Qeexo AutoML for Embedded AIAutomated Machine Learning Platform that builds tinyML solutions for the Edge using sensor data
Automotive
Mobile
IoT
QEEXO AUTOML: END-TO-END MACHINE LEARNING PLATFORM
Key Features Target Markets/Applications
For a limited time, sign up to use Qeexo AutoML at automl.qeexo.com for FREE to bring intelligence to your devices!
is for
building products
Automated Feature
Exploration and Model
Generation
Bill-of-Materials
Optimization
Automated Data
Assessment
Edge AI / TinyML
code for the smallest
MCUs
Reality AI Tools® software
Reality AI solutions
Automotive sound recognition & localization
Indoor/outdoor sound event recognition
RealityCheck™ voice anti-spoofing
info@reality.ai @SensorAI Reality AIhttps://reality.ai
SynSense builds ultra-low-power (sub-mW) sensing and inference hardware for embedded, mobile and edgedevices. We design systems for real-time always-on smart sensing, for audio, vision, IMUs, bio-signals and more.
https://SynSense.ai
Next tinyML Talks
Date Presenter Topic / Title
Tuesday,January 19
Lukas GeigerDeep Learning Researcher, Plumerai
Running Binarized Neural Networks on
Microcontrollers
Webcast start time is 8 am Pacific time
Please contact talks@tinyml.org if you are interested in presenting
Reminders
youtube.com/tinyml
Slides & Videos will be posted tomorrow
tinyml.org/forums
Please use the Q&A window for your questions
Andrew Baker
Andrew Baker joined Maxim Integrated in 2009. He has 25 years of experience in the electronics industry in roles ranging from development engineering to sales as well as business/product management. In his current role, he is responsible for leading Maxim’s wearable solutions initiatives for sensors and power management, as well as multiple other product lines. Andrew holds a Bachelor’s degree with honors in electronic engineering from the University of Portsmouth, UK.
The Future of Personalized Connected HealthcareAndrew Baker, Managing Director of Industrial & Healthcare Business Unit
Transitioning to a New Model for Healthcare Delivery
| Maxim Integrated 14
Global healthcare costs growing – Currently ~$9T or 10% of global GDP
Remote monitoring with analytics
Preventivemonitoring
Chronic disease management
Healthcare is Becoming More Personalized
• Medical/healthcare wearable devices totaled 640M in 2019
• Total device shipments forecasted to top 1B in 2023
• Target device types* set to grow at 22% CAGR (2019 to 2023)
| Maxim Integrated 15
Source: Omdia Healthcare Equipment Database, December 2019
*Select wearables include smartwatches, activity & fitness monitors, hearing aids, HRMs, disposable CGMs, infusion pumps and pregnancy test kitsResults are not an endorsement of Maxim Integrated. Any reliance on these results is at the third-party’s own risk.
Remote Patient Monitoring Use Cases – Virus Pandemic Model
Predictive Screening
Temp & SpO2
Onset Monitoring (Hi Risk)
Temp, SpO2, HR/ECG & Respiration
Periodic Telemetry
Post Hospital Monitoring
Temp, SpO2, Heart Rate/ECG, Respiration
24/7 Telemetry
Deteriorating Condition
Body Temperature, SpO2, Heart Rate/ECG, Respiration
24/7 Telemetry
ECG: ElectrocardiogramSpO2 : Blood oxygen saturation| Maxim Integrated16
Preventive Monitoring - AFib DetectionAtrial fibrillation (AFib) increases risk of stroke by 5x
| Maxim Integrated 17
Racing heart, fluttering or palpitations
Shortness of breath
Lightheadedness
People with no symptoms may be diagnosed by an exam and an ECG
Or no noticeable symptoms at allCommon symptoms of AFib
Source: American Heart Association
Chronic Disease Management - Continuous SpO2 MonitoringIdentify onset of critical conditions to mitigate risk of hospitalization
| Maxim Integrated18
251million
COPD cases globally; 5% of global deaths
Source: World Health Organization
Source: PubMed.gov
Source: THE LANCET
Source: BBC News
300million
Affected by Asthma globally
936million
30-69yrs adults with Obstructive Sleep
Apnea
48million
Infected with COVID-19 globally
$$$
Proven Track Record of Accelerating Time to Market for Customers
| Maxim Integrated19
Integrating several sensors for new functionality
Health Sensor PlatformMAXREFDES100# System Board
2016 2018 2020
Health Sensor Platform 2.0 (HSP 2.0)MAXREFDES101#
Health Sensor Platform 3.0 (HSP 3.0)MAXREFDES104#
Health Sensor Platform 3.0 (HSP 3.0)MAXREFDES104#: A wrist form-factor reference design
| Maxim Integrated 20
Clinical-grade
Accuracy meets regulatory
requirements for SpO2
& ambulatory ECG (IEC 60601-2-47)
Faster time to market
Saves at least six months in
development time
Complete reference
design
Source code and design files to
accelerate designs
Covers key vital signs
Addresses needs of advanced health
wearables with SpO2 , ECG, HR, HRV, RR, body
temp & motion
Detect infectious diseases, fever monitoringTemperature Trends
Critical Vital-Sign Measurements
| Maxim Integrated 21
Monitor pulmonary function, sleep disordersSpO2
Monitor respiration trendsRespiration Rate
Monitor heart rate trendsHeart Rate
AFib detection, cardiac healthECG
Vital Sign Use Cases
SpO2
HSP 3.0 Enables Clinical-Grade Use Cases
| Maxim Integrated 22
Heart rate
SpO2 & respiration
AFibdetection
Bodytemperature
prediction
Analytics (sleep,
stress, etc.)
HSP 3.0 System Block Diagram
| Maxim Integrated 23
Host MCU MAX32666(2xM4F CPUs
BLE4.2/5)
Algorithm HubMAX32670
(M4F)
Power Management IC
MAX20360
Flash Memory32MB
ECG + PPG AFEMAX86176
(110dB SNR, 120dB CMRR)
Temp SensorMAX30208
(±0.1oC Accuracy)
Accelerometer
Non-Maxim
Host Board Sensor Board
PC GUI(Windows 10)
• Raw PPG, ECG & Temp Data
• HR, SpO2
• ECG Waveform • Temp Output
Data Collection HSP 3.0
Maxim
Clinical-Grade Vital Sign Measurement Using PPG + ECG AFEMAX86176 enables the next generation of wearable healthcare use cases
• Synchronous acquisition of PPG & ECG measurements with independent sample rates
• Active Right Leg Drive (RLD) offers >110dB CMRR* for optimized ECG dry electrode performance
• Characterize ECG electrode material for system level optimization
• 110dB SNR* for highest performance SpO2
measurements
| Maxim Integrated24
PPG2-Ch Simultaneous acquisition
6x LED, 4x PD, Advanced Ambient Rejection
ECG1-Lead with Active RLD
AC/DC Leads Off Detection
Advanced FeaturesImpedance Measurement for ECG
electrode optimization
I/FSPI/I²C
FIFO256
Word
*CMRR: common mode rejection ratio; SNR: signal-to-noise ratio
The Wearable Healthcare Revolution: The Next Big Thing
Meeting Demands for Remote Patient Monitoring
Enabling Better Predictive/Preventive Healthcare & Chronic Disease Management
| Maxim Integrated 25
Maxim Enabling Personalized Healthcare
Copyright Notice
This presentation in this publication was presented as a tinyML® Talks webcast. The content reflects the opinion of the author(s) and their respective companies. The inclusion of presentations in this publication does not constitute an endorsement by tinyML Foundation or the sponsors.
There is no copyright protection claimed by this publication. However, each presentation is the work of the authors and their respective companies and may contain copyrighted material. As such, it is strongly encouraged that any use reflect proper acknowledgement to the appropriate source. Any questions regarding the use of any materials presented should be directed to the author(s) or their companies.
tinyML is a registered trademark of the tinyML Foundation.
www.tinyML.org